Featured Image from: Odyssey
By the beginning of the 1970’s, robotics had entered a stage of its life that was between infancy and intermediate development; people have moved past simply dreaming up of robotic humanoid servants and have actually begun developing prototypes.
Around that time, robotics professor Masahiro Mori noticed something very strange with people who came into contact with humanoid robots. Professor Mori noticed that robots that were constructed to look like robots rarely caused a reaction, although if the robot was of a particular novelty, it could cause positive reactions.
However, the more human a robot looked, the more negative the reaction from people. These reactions can range from an uncanny feeling of negativity to outright revulsion. But as robots approach an image that is indistinguishable from humans, those negative feelings disappear. Masahiro Mori called the phenomenon the “Uncanny Valley”, where “robotic” robots sit on one side of the valley and human-like androids in the other, with almost-human robots sitting in the valley.
As time went on, the Uncanny Valley became a concept that was adapted to computer graphics, wherein it was used to describe the uncannily negative reaction of people when they see a digital image that is photorealistic but not quite. This happens because the image fails to match up to our human perception of the world around us. One of the most famous examples of the Uncanny Valley’s affect on people is in the 2007 film Beowulf, which received criticism for its “unsettling” depiction of mythological monsters.
But that was 2007. In the past decade since then, advances in computer graphics has taken these computer-generated images into new heights. With this in mind, will the Uncanny Valley become a thing of the past in the near future? Or will the phenomenon take on new shape and function as it evolves with the technology that gave birth to it?
New Frontiers
Many analysts believe that the Uncanny Valley has already been conquered. Indeed, in the past decade alone, most CGI in movies and videogames—with depictions of weather, landscapes, buildings, planets, and various other background objects—has come to a point where it is almost, if not completely, indistinguishable from reality.
But that’s just the surroundings, what about humans? In the 2016 movie Rogue One, filmmakers used CGI to insert late actors Carrie Fisher and Peter Cushing into the film. While the feat was a technical marvel, many viewers were left unsettled by the jarring near-realistic depictions of the actors. Despite the hours poured in by animators into making the digital depictions as human as possible, it just wasn’t enough; many fans called it “weird”, even disrespectful, and is considered a blemish on an otherwise amazing movie.
Animator Alan Warburton predicts, however, that beyond conquering the Uncanny Valley, our society has entered a new frontier in computer graphics, with the Valley as just another region in an otherwise expansive map. Warburton’s analysis of computer graphics divides its evolution into four regions: the Uncanny Valley, the Wilderness, the Frontier, and the Beyond.
Warburton explains that his regions are both metaphysical and temporal spaces: both a period of time and a specific mindset for that time. The Uncanny Valley is a period that stretched sometime between its inception in the 70’s all the way to the present. During this period, multiple companies like Pixar and Blue Sky, to name a couple, developed new technologies to model, animate, and ultimately simulate as many aspects of our reality as possible, from blades of grass and raindrops to animal fur and human hair.
While these technologies were mostly developed for films, videogames, or even as showcases for conferences, they eventually found their way into commercial software packages that gave everyone the power to create their own digital realities. Warburton calls this period “the Frontier”, where individuals are free to create computer-generated images and videos, and even videogames for the dedicated nerds, that can rival the output of major studios. This, he argues, forces companies to up their game and create bigger, better, and more realistic graphics that will blow the competition out the water.
But what happens when we produce computer-generated images and videos that are beyond photorealistic? What if we create a future that is dominated by CGI that is so real, people can no longer discern fact from fiction?
Beyond the Valley
This post-truth phenomenon is what Warburton calls “the Beyond”, a point in time where extremely advanced CGI can create a paradoxical effect where the more realistic an image looks, the less real people perceive it to be.
This becomes problematic when political forces leverage CGI to forward their nefarious ends: soon enough, hyper realistic CGI can be used to manipulate images and videos to show political opponents doing or saying things that are far from reality. What’s worse is that, because of the advanced quality of the CGI, people will tend to believe it.
But analysts like Warburton believe that future developers will find safeguards against this. He and others like him like to focus more of what “the Beyond” can offer, mainly, something he calls “theoretical photorealism”. Theoretical photorealism is the use of CGI to render heretofore unseen images, like how animators collaborated with Physicists and other scientists to create as accurate a rendering of a black hole possible for the 2014 film Interstellar, a move that was heavily praised by the scientific community.
Ironically, this rendition of a black hole was immediate accepted by the general public as “reality”, despite the fact that no one has ever captured an image of a black hole in real life (all depictions have been based on mathematical formulas). This further illustrates Warburton’s “the Beyond”, a point in time where CGI can be considered reality.
But beyond even “the Beyond” lies “the Wilderness”, a vast and seemingly unregulated space wherein animators, consumers, and individual creators can focus not on realism, but on creating images and videos and games that are meant to entertain. Already, Pixar is doing this with its focus on its cartoonish
